direct inhibition of in vitro pld activity by 4-(2-aminoethyl)-benzenesulfonyl fluoride

D4

BA*W

R

ltpiaflvntAgcpap(ediP

n

Pbpsbaacbsm

Biochemical and Biophysical Research Communications 273, 302–311 (2000)

doi:10.1006/bbrc.2000.2938, available online at http://www.idealibrary.com on

0CA

irect Inhibition of in Vitro PLD Activity by-(2-Aminoethyl)-Benzenesulfonyl Fluoride

rooke Andrews,* Kristina Bond,* Jason A. Lehman,* Jeffrey M. Horn,*,†my Dugan,*,† and Julia Gomez-Cambronero*,1

Department of Physiology and Biophysics and †Department of Anatomy,right State University School of Medicine, Dayton, Ohio 45435

eceived May 23, 2000

Phospholipase D catalyzes the hydrolysis of phos-pThgltma

wSrpatnnsPwsnp

iiovpcaeciiPa

While conducting a purification protocol of phospho-ipase D (PLD) from human granulocytes, we observedhat PLD activity was inhibited by a commonly-usedrotease inhibitor cocktail. Of the six inhibitors present

n the cocktail, the serine protease inhibitor, 4-(2-minoethyl)-benezensulfonyl fluoride (AEBSF), wasound to be the sole inhibitor of PLD. AEBSF caused aoss of neutrophil and purified plant PLD activities initro, but not in intact cells at the concentrations used,or did it affect the related phospholipases A2 and C,hat were utilized as specificity controls. The compoundEBSNH2, which has the fluoride replaced by an -NH2

roup, failed to affect PLD activity as did otherompounds structurally related to AEBSF with knownrotease inhibitory capabilities. Finally, basal- andgonist-stimulated PLD activity was inhibited in phos-hatidylcholine-specific anti-PLD immunoprecipitatesIC50 5 75 mM). These results suggest that AEBSF, in anffect probably unrelated to its anti-proteolytic ability,irectly interferes with PLD enzymatic activity, making

t a significant compound to begin analyzing the role ofLD in mammalian cell signaling. © 2000 Academic Press

Key Words: PLD; neutrophils; protease inhibitor; sig-al transduction.

Abbreviations used: PLD, phospholipase D; PC, phosphatidylcholine;A, phosphatidic acid; PEt, phosphatidylethanol; PBu, phosphatidyl-utanol; PIP2, phosphatidylinositol-4,5-bis-phosphate; PC8, dioctanoylhosphatidylcholine; PIC, protease inhibitor cocktail; E-64, trans-epoxy-uccinyl-1-leucylamido-(4-guanidino) butane; AEBSF, 4-(2-aminoethyl)-enezensulfonyl fluoride; AEBSNH2, 4-(2-aminoethyl)-benzene-sulfon-mide; TLCK, tosyl-lysine chloromethyl ketone (N-[5-amino-1-(chloro-cetyl)pentyl]-4-methyl-benzenesulfonamide); TPCK, tosyl-phenylalaninehloromethyl ketone (N-[3-chloro-2-oxo-1-phenylmethyl)propyl]-4-methyl-enzenesulfonamide); pTF, p-toluenesulfonyl fluoride; pTCl, p-toluene-ulfonyl chloride; PMA, phorbol, 12-myristate, 13-acetate; MAPK,itogen-activated protein kinase; FMLP, fMet-Leu-Phe.1 To whom correspondence should be addressed.

302006-291X/00 $35.00opyright © 2000 by Academic Pressll rights of reproduction in any form reserved.

hatidylcholine into choline and phosphatidic acid.he wide range of fates that phosphatidic acid (PA)as, enables it to become a significant second messen-er and ties phospholipase D to many important cellu-ar events including: mitogenesis, cell signaling, syn-hesis of superoxide radicals, inflammation, exocytosis,embrane delivery, vesicle budding, phagocytosis, and

poptosis [reviewed 1–11].There are many pathways involved in PLD signalingith many upstream and downstream regulators.ome of the proposed regulators are the protein ty-osine kinases [1, 6–8, 10] and the mitogen-activatedrotein kinases (MAPKs), which may act as upstreamnd/or downstream regulators of PLD [12–16]. Both ofhese types of kinases are very abundant in the humaneutrophil and could likely facilitate PLD activation. Aumber of inhibitors of PLD activity have been de-cribed; however, they generally do not act directly onLD itself. Rather, they cause inhibition by interferingith molecules that are required for PLD activation,

uch as the small GTP-binding proteins [17–19, 21],amely ARF and/or Rho, PIP2 [17, 21], tyrosine phos-horylation [20, 22, 23] or lipid repression [26].We have uncovered that neutrophil PLD activity is

nhibited in the presence of a commonly-used proteasenhibitor cocktail. The common and unexplainable lossf PLD activity during protein purification observed byarious authors, might be explained if a certain com-onent of the protease mix used in cell disruption andolumn chromatography buffers were to inhibit PLDctivity when it is later assayed from eluates. Theffect that each individual protease inhibitor in theocktail had on PLD activity was analyzed separatelyn order to determine which specific inhibitor or inhib-tors were responsible for the observed decrease inLD activities. The serine protease inhibitor 4-(2-minoethyl)-benzenesulfonyl fluoride, AEBSF, was

found to be the only inhibitor of neutrophil PLD activ-idptmspf

M

pr(Awfm((clatTtTfampaCafem

pmfcmapfNlflucimp

F8tc

placed on ice for 10 min to promote further lysis. Protein concentra-ttfP

adsmt4Tmcwcbwbab

odBtcnbm4cav

rSof31sstggD

difusaftPpsit

tt

Vol. 273, No. 1, 2000 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

ty and the effect was exerted directly on the enzyme asemonstrated in three cell-free systems: cell sonicates,urified plant enzyme, and anti-PLD immunoprecipi-ates. Although the IC50 for the latter in vitro enzy-atic assays was 75 mM, AEBSF only represents a

tarting point for structural modification that couldotentially evolve into a much-needed specific inhibitoror studying PLD signaling.

ATERIALS AND METHODS

Materials. 1,2-Dioctanoyl-sn-glycero-3-phosphocholine (PC8), L-a-hosphatidyl-choline, b-arachidonoyl-g-stearoyl (arachidonoyl-PC), pu-ified cabbage PLD, bee venom PLA2, pancreatic PLA2 and silica gel70-230 mesh, 60 Å pore), the protease inhibitor cocktail (P-2714),EBSF, EDTA, bestatin, E-64, leupeptin, aprotinin, PMA and FMLPere obtained from Sigma (St. Louis, MO); Bacillus cereus PLC was

rom Boehringer Mannheim (Indianapolis, IN); n-[1-3H] butanol (5 Ci/mol) and L-a-phosphatidylcholine, b-arachidonoyl [3H]-g-stearoyl PC

200 Ci/mmol) were purchased from American Radiolabeled ChemicalsSt. Louis, MO); L-3-phosphatidylcholine, 1-stearoyl-2-[1-14C]ara-hidonyl-3-PC (114 mCi/mmol) was from Amersham Life Science (Ar-ington Heights, IL); 1,2-dioleoyl-sn-glycero-3-phosphoethanol (PEt)nd 1,2-dioleoyl-sn-glycero-3-phosphobutanol (PBu) standards were ob-ained from Avanti Polar Lipids (Alabaster, AL); LK6D silica gel 60 ÅLC plates were from Whatman (Clifton, NJ); Scintiverse II scintilla-ion cocktail was purchased from Fisher (Pittsburgh, PA); AEBSNH2,LCK and TPCK were from RBI (Natick, MA); pTF and pTC1 were

rom Aldrich (Milwaukee, WI); electrophoresis and Bradford proteinssay chemicals were from Bio-Rad Laboratories (Richmond, CA); Im-obilon PVDF membranes were from Millipore (Bedford, MA). Anti-

hosphotyrosine (PY-20 clone) affinity purified monoclonal antibodynd anti-p42-MAPK monoclonal antibody were purchased from Santaruz Biotechnology (Santa Cruz, CA); anti-PC specific PLD1 andnti-PC specific PLD2, N-terminal, affinity purified antibodies wererom QCB (Camarillo, CA). Enhanced chemiluminescence (ECL) West-rn blotting detection reagents were obtained from Amersham Phar-acia Biotech (Piscataway, NJ).

Isolation of human peripheral blood neutrophils. The isolationrocedure was based on English and Anderson [27], with someinor modifications. Between 50 –55 ml of blood were collected

rom the antecubital vein of healthy individuals in a 60-ml syringeontaining 6 ml of sodium citrate as anticoagulant. Blood wasixed with 15 ml of 6% dextran, allowed to settle, and the plasma

nd buffy coat were removed and spun down at 800g for 5 min. Theellet was resuspended in 35 ml of saline and centrifuged at 800gor 15 min at 10°C in a Ficoll-Histopaque discontinuing gradient.eutrophils were recovered and contaminating erythrocytes were

ysed by hypotonic shock. Cells were spun again and the pellet wasash frozen in dry ice and ethanol and stored at 270°C until laterse. Since no heparin was used as an anticoagulant, but sodiumitrate instead, the possibility of activation during isolations extremely small. No neutrophil aggregation (i.e., the hall-

ark for neutrophil activation) was observed during this isolationrocedure.

Preparation of cell sonicates and addition of protease inhibitors.lash-frozen cells were allowed to thaw on ice and 5 mM HEPES pH.0 (hypotonic sonication buffer) was added in differing concentra-ions based upon pellet size. The cells were resuspended and soni-ated on ice (three cycles, 10 sec each). The sonicated cells were then

303

ion was determined as previously described by Bradford [28] andhe sonicates were kept on ice until used in PLD activity assays, orrozen immediately after stimulation for activity assays or SDS–AGE.

Measurement of PLD and other phospholipases. Phospholipase Dctivity was directly measured in vitro by utilizing the transphosphati-ylation reaction with short-chain PC (PC8) and [3H]butanol exogenousubstrates as previously described by Davis et al. [29] with only minorodifications. 50 ml of cell sonicates (0.8–1.2 mg/ml protein) were added

o 1.5 ml-Eppendorf tubes containing 50 ml of the following assay mix:.5 mM phospholipid, 75 mM HEPES, pH 7.9, and 4.7 mCi [3H]butanol.he resulting PLD reaction (final volume 100 ml) was incubated for 15in at 30°C. The reactions were stopped by adding 0.3 ml ice-cold

hloroform/methanol (1:2) and 70 ml of 1% perchloric acid. The tubesere then vortexed and 100 ml of 1% perchloric acid and 100 ml of

hloroform were added. The tubes were again vortexed and centrifugedefore aspirating the upper phase. The lower phase was washed onceith 100 ml 1% perchloric acid and an aliquot of 125 ml was removed toe dried for thin-layer chromatography (TLC). TLC lanes were scrapednd dissolved in 3 ml of Scintiverse I scintillation cocktail and countedy scintillation spectrometry.Phospholipase A2 activity was measured in vitro by the production

f radiolabeled arachidonic acid from an exogenous substrate asescribed by Van den Bosch et al. [30] with minor modifications.riefly, samples (neutrophil sonicates or purified PLA2) were added

o 1.5 ml-Eppendorf tubes containing the following final reactiononditions (in 100 ml): 0.25 M Tris–HCl, pH 8.5, 10 mM CaCl2, 12mol arachidonoyl-PC and 0.2 mCi [3H]arachidonoyl-PC, and incu-ated for 20 min at 30°C. The reactions were stopped by adding 0.5l Dole’s extraction medium (2-propanol, n-heptane, 1 N H2SO4;

0:10:1, v/v), 0.4 ml n-heptane and 0.3 ml H2O. The upper phase washromatographed over a 240-mg, 70-230 mesh, silica gel column and0.6-ml diethyl ether wash was collected directly into scintillation

ials and counted by scintillation spectrometry.Phospholipase C activity was measured in vitro by the production of

adiolabeled diglycerides from an exogenous substrate as described bymith and Waite [31] with some modifications. Samples (cell sonicatesr purified PLC) were added to 1.5 ml-Eppendorf tubes containing theollowing final reaction conditions (in 100 ml): 15 mM Tris–HCl, pH 7.4,0 mM NaCl, 1.5 mM CaCl2, 50 nCi L-3-phosphatidylcholine,-stearoyl-2-[1-14C]arachidonyl-3-PC, 12 nmol distearoyl-PC and 1 mModium deoxycholate, and incubated for 10 min at 30°C. Lipids wereeparated by TLC in benzene/chloroform/methanol (80:15:3.25; v/v/v)hat separates DAG subspecies (Rf ; 0.4) from PA (Rf 5 0), triacyl-lycerides (Rf ; 0.7) and other neutral lipids (Rf ; 0.9). Zones comi-rating with authentic standards (C14:0/C14:0-DAG; C16:0/C18:0-AG) were scraped and counted.

In vitro and intact-cell enzymatic assays. The standard proce-ure used in this paper for PLD, referred to as “in vitro” assaynvolved neutrophil cells that were flash frozen upon isolationrom human blood, thawed, sonicated immediately and left on icentil AEBSF was added, to either the assay mix or to the cellonicates (but not to both), with a similar outcome, at the time ofssaying enzymatic activity. For an “intact cell” stimulation,reshly isolated neutrophils (1 3 107 cells/ml cells) were pre-reated with AEBSF for 15 min and then stimulated with eitherMA (50 ng/ml) for 5 min or with fMet-Leu-Phe (100 mM) inresence of cytochalasin B (0.1 mg/ml) for 3 min. The cells werepun down (13,000g, 15 sec) to remove inhibitor (unless otherwisendicated) and pellets were flash frozen. Later, the cells werehawed, sonicated and assayed for activity.

Immunoprecipitation and Western blotting. For immunoprecipi-ation, total cell sonicates were incubated with the immunoprecipi-ating anti-PLD antibodies [32]. PC-specific anti-PLD antibodies

wrEmm1wfic41iotnrr

RESULTS

E

tti(sTpTittactptaca

D

tsc

TABLE 1

AEBELA

oas

war


ere conjugated overnight with goat anti-rabbit IgG linked to aga-ose beads in a conjugation buffer (10 mM HEPES, pH 7.3, 1 mMGTA, 0.1 mM PMSF, 0.21 mM sodium orthovanadate, 1 mM am-onium molybdate, 1.2 mM diisopropyl fluorophosphate (DFP), 10M p-nitro-phenyl-phosphate (PNPP), 0.5% Triton X-100, and 0.5

00 mg/ml each of leupeptin, aprotinin, and pepstatin A). Conjugatesere thoroughly mixed (1:1; v/v) with cell sonicates for 4-h (antibodynal concentration, 3 mg/ml). Immune complexes were recovered byentrifugation and washed with buffer A (100 mM Tris–HCl, pH 7.4,00 mM LiCl) and two times with buffer B (10 mM Tris–HCl, pH 7.4,00 mM NaCl, 1 mM EDTA). Immune complexes were resuspendedn a final volume of 60 ml with sonication buffer and assayed for PLDr with 60 ml of sonication buffer for subsequent PLD assays underhe same conditions described above. Controls were run with immu-oprecipitates carried in the absence of primary antibodies andesulting background radioactive counts were subtracted from testesults.

Protease Inhibition Cocktail (PIC) Composition

PIC Concentration

23 13 0.53 0.253

EBSF (mM) 2 1 0.5 0.25DTA (mM) 1 0.5 0.25 0.12estatin (mM) 130 65 32.5 16.25-64 (mM) 1.4 0.7 0.35 0.17eupeptin (mM) 1 0.5 0.25 0.12protinin (mM) 0.3 0.15 0.075 0.037

Note. The PIC lyophilized powder (1 bottle) was dissolved in 1 mlf H2O to prepare a 1003 stock from which aliquots were taken tochieve the indicated dilutions in the PLD in vitro assay, as de-cribed under Materials and Methods.

304

ffects of Protease Inhibitor Cocktail (PIC)on Neutrophil PLD Activity in Vitro

After noticing a great loss of enzyme activity alonghe different steps of protein purification, we suspectedhat the column buffers might contain an inhibitor. Wenvestigated whether a protease inhibitor cocktailPIC) (a mixture of water-soluble inhibitors with broadpecificity for Ser, Cys, Asp and metallo-proteases; seeable 1 for composition]) routinely included in proteinurification had an effect on PLD activity in vitro.hus, we incorporated PIC directly into the PLD activ-

ty assay described under Methods and we observedhat it indeed inhibited the phospholipase activity athe two pH’s (in HEPES buffer) measured (Fig. 1A)lthough slightly more at the more basic pH. Next, wearried out activity assays using HEPES (pH 8.0, op-imal) and a range of PIC concentrations. The dataresented in Fig. 1B shows that addition of the pro-ease inhibitor cocktail is inhibitory to PLD activity in

concentration-dependent manner, with 0.5 3 PIConcentration resulting in ;50% inhibition of PLDctivity.

etermination of the Specific Compound(s)Responsible for PLD Inhibition

In order to determine the specific protease inhibi-or(s) responsible for inhibiting PLD activity, we as-ayed PLD in the presence of each of the six individualomponents of the cocktail at varying concentrations:

FIG. 1. Effects of PIC on neutrophil PLD activity in vitro. (A) PLD activity assays were carried out in HEPES buffer at the indicated pHsith (solid bars) or without (hatched bars) PIC. (B) Various concentrations of PIC were incorporated into the assay mix as indicated andctivity assays were conducted as described (see Table 1 for concentration of individual inhibitors at the indicated dilutions). These dataepresent the average 6 SEM of four separate experiments each done in duplicate.

pccb


FIG. 2. Effects of each individual PIC protease inhibitor on in vitro neutrophil PLD activity. PLD activity was assayed separately in theresence of each individual protease inhibitor contained within the cocktail: EDTA, AEBSF, aprotinin, leupeptin, bestatin, and E-64. Variousoncentrations (1/83 through 23 PIC concentration) of each inhibitor were incorporated into the assay mix, assayed, and compared toontrols as shown. The data represent the average 6 SEM of four separate experiments each done in duplicate. Horizontal dotted lines haveeen drawn at control values.

305

Em(oCbpP

P

pttsptb(a

A

tctitt

pretreated with 500 mM AEBSF for 15 min and thenslusta;smTtiwd(

A

efaroPApt(

ceiaiae

cppac


DTA (0–2 mM), AEBSF (0–4 mM), aprotinin (0–0.5M), leupeptin (0–2 mM), bestatin (0–260 mM), or E-64

0–2.8 mM). AEBSF was identified as the sole inhibitorf PLD activity, having an IC50 of ;50 mM (Fig. 2).onversely, EDTA and leupeptin showed no inhibition,ut a slight stimulatory effect. This suggests that theserotease inhibitors are perhaps somewhat protective ofLD activity.

urified Plant PLD

Similar activity assays were conducted in vitro withurified cabbage phospholipase D in place of the neu-rophil sonicates. This was done in order to examinehe effects of AEBSF on PLD in the absence of anyerine proteases and other molecules that are usuallyresent in unpurified samples. Figure 3 shows thathis purified cabbage PLD is also inhibited by AEBSF,ut to a lesser (;25%) extent than neutrophil PLD75%). It also indicates that AEBSF can inhibit PLD bydirect mechanism.

EBSF on in Vitro and on Intact-Cell PLD Activities

AEBSF at various concentrations was included inhe in vitro PLD enzyme assay mixture. Figure 4 indi-ates that AEBSF exerts its inhibitory effect at shortimes (;20% activity loss at 0.5 min) and that 250 mMs needed for statistically significant differences in ac-ivity. A different set of experiments was also under-aken where freshly isolated, intact, neutrophils were

FIG. 3. Effects of various concentrations of AEBSF on purifiedabbage PLD in vitro. Various concentrations of AEBSF were incor-orated into the assay mix and activity assays were conducted usingurified cabbage PLD as the enzyme source. The data represent theverage 6 SEM of three separate experiments each done in dupli-ate. The horizontal dotted line has been drawn at control values.

306

timulated with PMA or FMLP. In the first set (Fig. 5,eft), cells were washed free off the inhibitor after stim-lation. AEBSF did not inhibit basal or agonist-timulated PLD activity. When AEBSF was present athe time of cell sonication (Fig. 5, middle), basal andgonist-stimulated PLD activities were lessened by60%. Further, when AEBSF was present during cell

onication and at the time of PLD activity measure-ent (Fig. 5, right), the inhibition reached ;85%.hus, the inhibitory effect of AEBSF is observed at theested concentrations in the in vitro setting but not inntact cells. A lack of effect in whole-cell neutrophilsas observed on two common routes of signal trans-uction: tyrosine phosphorylation and p42-MAPKFig. 6).

EBSF on PLD Immunocomplexes

To ascertain if AEBSF would inhibit PLD in a morenriched sample than whole cell sonicates, we per-ormed immunoprecipitation with specific anti-PLDntibodies and then assaying the immunocomplex aga-ose beads for PLD activity in the presence or absencef AEBSF. The results presented in Fig. 7 indicate thatLD activity was strongly inhibited by AEBSF in vitro.calculated IC50 of ;75 mM in this enriched sample

reparation was a value close to one order of magni-ude higher than that observed for whole cell sonicatesFig. 2).

FIG. 4. Time course/dose dependency of AEBSF. Differing con-entrations of AEBSF were added to cell sonicates at the time ofnzyme assay and incubated at 30°C for the times shown. After thendicated time, the sonicates were removed from incubation andssayed for activity. For reliable comparisons, each sample wasncubated for exactly 20 min in the activity assay. The results are theverage of three separate experiments. The data represent the av-rage 6 SEM of three separate experiments.

E

PrsectBA

S

siwpgpeA

D

o

rification. Without them, uncontrolled proteolysiswlttpidc(nawt

Mn(apr

c5sbdifla(stmd


ffect on Other Phospholipases

We next examined the effect of AEBSF on relatedC-specific phospholipases, namely, PLA2 and PLC. Aseported in Table 2 no appreciable inhibition was ob-erved on PLA2 from neutrophil sonicates or purifiednzyme from either bee venom or from bovine pan-reas. Similarly no inhibition was present on PLC ac-ivity from either neutrophil or purified enzyme fromacillus cereus. Thus, the observed inhibitory effect ofEBSF is specific for PLD activity.

tructural Analogues of AEBSF

Finally we aimed to investigate what part of thetructure in the AEBSF molecule was responsible fornhibition. Out of the several structural analogues thatere considered (Table 3), PLD inhibition was mostotent in AEBSF (Table 4). Substitution of the fluorideroup for -NH2 abrogated the inhibitory effect. The-aminoethyl group was also important to confer anffect. Quantitatively, the inhibitory potency was:EBSF @ TLCK . pTF . pTCl @ TPCK 5 AEBSNH2.

ISCUSSION

It has been known for a long time that the presencef protease inhibitors is a requirement for protein pu-

FIG. 5. Effects of AEBSF on neutrophil PLD activity in intactells and in vitro. Neutrophils (1 3 107 cells/ml) were pretreated with00 mM AEBSF for 6 min at 37°C. Following this incubation,ome cells were stimulated with PMA or FMLP for 5 min (hatchedars) or left untreated (solid bars). After stimulation, cells wereivided in three sets and spun down (13,000g, 15 sec) to remove thenhibitor, the supernatant was then decanted and the pellets wereash frozen. The first set (left group of bars) was sonicated in thebsence of AEBSF and the second (middle group of bars) and thirdright group of bars) sets were sonicated in its presence. Aliquots ofonicates were assayed for PLD activity in the standard assay mixhat further included AEBSF only for the third set. These experi-ents are the average 6 SEM of three separate experiments each

one in duplicate.

307

ould occur and result in considerable damage to cel-ular proteins and an abundant loss in enzyme func-ionality. In our search for an appropriate concentra-ion of inhibitors to use for mammalian PLDurification, we found that a commonly used proteasenhibitor cocktail (PIC) resulted in a concentration-ependent inhibition of neutrophil PLD activity fromolumn purification eluates. Only a 4 times dilution0.253) of the original concentrate allowed for mainte-ance of higher protein levels by preventing proteolysisnd preserving PLD activity. We report that AEBSFas highly inhibitory while the other protease inhibi-

ors in the cocktail, at the manufacturer’s recom-

FIG. 6. Effects of AEBSF on neutrophil PY-proteins and p42-APK. Western blots were prepared from whole cell sonicates after

eutrophils were stimulated with either 200 mM (A) or 1 mM AEBSFB) for the indicated lengths of time. (C, D) Represent the same blotsfter they were stripped of anti-phosphotyrosine antibodies and re-robed with anti-p42-MAPK (ERK2) antibodies. These results areepresentative of two different experiments.

mewsits2t

a

ssoFbsmtlthmKlf

Nwleateace

LE

p[ppi

TABLE 3


ended concentrations, showed little or no significantffects on PLD activity. It is worth noting that AEBSForking concentrations [EC50 5 500 mM in the in vitro

tudies (Fig. 2) and 75 mM for the enzyme-enriched/mmunoprecipitation setting (Fig. 7)] are relatively lowo achieve an effect on intact cells (Fig. 5). In previoustudies, authors have used 2 mM or 200 mM for a4-hour incubation to achieve AEBSF-induced inhibi-ion on the NADPH oxidative burst [24, 33–35].

As for the exact mechanism by which a serine ester-se inhibitor exerts its effects on PLD, several points

FIG. 7. Effect of AEBSF on PLD activity in immunocomplexes.eutrophil cells were subjected to stimulation with FMLP for 3 min,ashed flash-frozen, and sonicated. The cell sonicates (approximate-

y 1.6 mg protein/ml) were subjected to immunopreciptation withither 3 mg/ml anti-PLD1 or anti-PLD2 affinity-purified polyclonalntibodies. Immunocomplex agarose beads were recovered by cen-rifugation and assayed for in vitro PLD activity with [3H]butanol aslaborated under Materials and Methods in the presence or thebsence of the indicated concentrations of AEBSF. Immunopre-ipitation data presented are the average 6 SEM of four separatexperiments.

TAB

Effect of AEBSF Inhibito

PLA2 activity

NeutrophilBee venom(60 units)

Control 874 6 55 17603 6 1046100 mM 890 6 70 19537 6 957250 mM 851 6 59 17204 6 938500 mM 803 6 65 18365 6 1224

Note. Phospholipase activities were measured in vitro against anresence or absence of the indicated concentrations of AEBSF. Res

14C]-diacylglycerol dpm (PLC activity). “Neutrophil” refers to 50 mlurified phospholipases were: PLA2 from been venom, 600–1800 unrotein at pH 8.0; PLC from Bacillus cereus, 2000 units/mg protein atndependent experiments performed in quadruplicate.

308

hould be considered. First, does it work on an up-tream mechanism that regulates PLD in the processf cell signaling or does it directly affect the enzyme?or the former, superoxide release has been shown toe prevented and/or inhibited by AEBSF in variousystems such as macrophages and monocytes [35]. Aechanism was proposed in which the enhanced func-

ions of priming for superoxide release could be regu-ated by proteolysis by serine proteases, although thearget protein of the proteases was unknown. Othersave proposed the possibility of such a proteolyticechanism for PLD activation in response to inhibitor.essels et al. described the inhibitor of chymotrypsin-

ike proteases, zLYCK, in human neutrophils withMLP-induced PLD [21]. Also, Lukowski et al. suggest

2

n Other Phospholipases

PLC activity

Pancreas(3 units) Neutrophil

B. cereus(100 units)

3416 6 212 1066 6 66 72879 6 40303088 6 249 1190 6 79 68715 6 36413509 6 199 1392 6 100 74933 6 48194127 6 260 1122 6 91 77456 6 5111

propriate exogenous substrate (see Materials and Methods) in thes shown are free [3H]-arachidonic acid dpm (PLA2 activity) or freeeutrophil cell sonicates (0.8–1.2 mg protein/ml). Original stocks of

mg protein at pH 8.9; PLA2 from bovine pancreas, 25–75 units/mg6.6–8.0. Values presented in the table are the mean 6 SEM of three

Chemical Structures of AEBSF Analogs Usedin This Study

r o

apultof nits/pH

toa3iasis

otimiwinataPbcaats

p(tCasfP

PePTpwaiHaaplaTPiipio

umPsstmPs

A

pf

TABLE 4

twTS


hat the protein fodrin could be involved in regulationf cytoskeletal proteins and PLD by a proteolytic mech-nism [25] or by blocking access of PLD to PIP2 [17,6]. Inhibition of PLD activity has also been describedn the presence of Rho-GDP dissociation inhibitor [18],nd presqualene diphosphate (PSDP) is the first lipidicubstance of intracellular origin that shows a directnhibition of PLD and LTB4- and lipoxin A4-stimulateduperoxide release [26].However, it is our belief that AEBSF works directly

n the enzyme rather than on an upstream mechanismhat regulates PLD, based on the following: (i) AEBSFnhibits PLD in vitro in a dose- and time-dependent

anner, and at the concentrations used, does not affectntact cells, whether they were resting or stimulatedith classical neutrophil agonists; (ii) AEBSF inhib-

ted purified plant PLD; (iii) AEBSF inhibited immu-oprecipitated PLD (i.e., a high-enriched enzyme prep-ration). However, it can not be excluded at this timehe possibility that AEBSF inhibits PLD activity byltering an upstream regulator (a serine protease) ofLD that had been coimmunoprecipitated by the anti-ody. This possibility seems unlikely since immuno-omplexes treated with a glycine-based buffer (pH 5 2)nd neutralization to pH 5 7.4, to dissociate thentigen-antibody complex still showed the same pat-ern of activity and/or activity inhibition as thosehown in Fig. 7.If a direct effect of AEBSF on PLD is to be contem-

lated, it remains to be learned the exact mechanismi.e., catalytic or allosteric site) by which a serine es-erase inhibitor exerts its effects on a phospholipase.hemical alterations on the molecule will undoubtedlyid in answering this question. The study using the 5tructural analogues (Tables 3 and 4) revealed severalacts regarding the chemical properties needed to elicitLD inhibition. The fluoride atom is necessary for its

Effect of Structural Analog

A

0 100

AEBSF 31981 6 1599 29110 6AEBSNH2 30252 6 1510 30192 6TLCK 29756 6 2289 28473 6TPCK 27397 6 1917 28796 6pTF 34103 6 1364 33721 6pTCl 31493 6 1420 29981 6

Note. Neutrophil PLD activity ([3H]butanol dpm/mg protein) wasrations of AEBSF and 5 structural analogs. To normalize the data (dater insoluble), all 6 were dissolved as concentrated stocks in DMShe final concentration of DMSO was fixed at 0.5 ml/100 ml in each caEM of four independent experiments performed in duplicate.

309

LD inhibitory effect, as demonstrated by the lack offfectiveness by AEBNH2. Some marginal inhibition ofLD was observed with the chloromethyl ketoneLCK, but not with TPCK, that has an extra hydro-hobic benzene ring. Some inhibition was observedith pTF indicating that the p-aminoethyl moiety islso key for the PLD inhibitory effect perhaps by mak-ng a nucleophilic attack to the substrate more likely.owever, the observation that AEBSNH2 (which islso inactive in inhibiting proteolysis) resulted in anbsence of PLD inhibition, does not mean that anyrotease inhibitor will be a PLD inhibitor. In fact,eupeptin, aprotinin, bestatin, E-64, and the structuralnalogues (that are also protease inhibitors) TLCK,PCK, pTF and pTCl had marginal or no effect onLD, which led us to hypothesize that AEBSF might

nhibit PLD without involving its antiproteolytic activ-ty per se. In this vein, AEBSF has been describedreviously as inhibiting activation of macrophages bynhibition of signal transduction rather than inhibitionf proteases [37].Finally, a systematic study of this compound allowed

s to conclude that AEBSF was unable to exert anyeasurable effect on related phospholipases, namelyLA2 and PLC (Table 2), further establishing its pos-ible usefulness as a specific inhibitor of PLD. Theynthesis of a compound based on the AEBSF struc-ure that would act in vitro at nM concentrations wouldake this inhibitor useful to explore a putative role ofLD in cell signaling in an investigator’s particularystem of study.

CKNOWLEDGMENTS

The authors thank Dr. Dan Halm for fruitful discussions on thisroject and critical reading of the manuscript, and Katie Ruetenikor her experimental assistance. This work has been supported in

f AEBSF on PLD Activity

log Concentration (mM)

250 500

92 25757 6 1152 17127 6 119821 28829 6 1960 28446 6 227063 27012 6 1160 24589 6 170715 31122 6 2391 30239 6 211683 30740 6 1816 29345 6 140441 30184 6 12475 29028 6 2177

asured in vitro in the presence or absence of the indicated concen-to the fact that some compounds were water soluble and others werend added to the PLD reaction mixture (see Materials and Methods).DMSO vehicle alone added to 0 mM column). Values are the mean 6

s o

na

181720201723

meueO ase (

part by the National Institutes of Health Grant HL056653 andAf

R

1

1

1

1

1

1

1

1

18. Bowman, E. P., Uhlinger, D. J., and Lambeth, J. D. (1995)

1

2

2

2

2

2

2

2

2

2

2

3

3

3

3

3


merican Heart Association 9806283 (J.G.-C.). Kristina Bond was aellow of the STREAMS Program (NIH #HL07805).

EFERENCES

1. Spiegel, S., Foster, D., and Kolesnick, R. (1996) Signal transduc-tion through lipid second messengers. Curr. Opin. Cell. Biol. 8,159–167.

2. Feig, L. A., Urano, T., and Cantor, S. (1996) Evidence for Ras/Ralsignaling cascade. Trends Biochem. Sci. 21, 438–441.

3. Cockcroft, S. (1996) Phospholipid signaling in leukocytes. Curr.Opin. Hematol. 3, 48–54.

4. Roth, M. G., and Sternweis, P. C. (1997) The role of lipid signal-ing in constitutive membrane traffic. Curr. Opin. Cell. Biol. 9,519–526.

5. Gomez-Cambronero, J., and Keire, P. (1998) Phospholipase D: Anovel major player in signal transduction. Cell. Signal. 10, 387–397.

6. Houle, M. G., and Bourgoin, S. (1999) Regulation of phospho-lipase D by phosphorylation-dependent mechanisms. Biochim.Biophys. Acta 1439, 135–149.

7. McPhail, L. C., Waite, K. A., Regier, D. S., Nixon, J. B.,Qualliotine-Mann, D., Zhang, W. X., Wallin, R., and Sergeant, S.(1999) A novel protein kinase target for the lipid second messen-ger phosphatidic acid. Biochim. Biophys. Acta 1439, 277–290.

8. Exton, J. H. (1999) Regulation of Phospholipase D. Biochim.Biophys. Acta 1439, 121–133.

9. Jones, D., Morgan, C., and Cockcroft, S. (1999) Phospholipase Dand membrane traffic. Potential roles in regulated exocytosis,membrane delivery and vesicle budding. Biochim. Biophys. Acta1439, 229–244.

0. Lennartz, M. R. (1999) Phospholipases and phagocytosis: Therole of phospholipid-derived second messengers in phagocytosis.Int. J. Biochem. Cell Biol. 31, 415–430.

1. Lacal, J. C. (1997) Regulation of proliferation and apoptosis byRas and Rho GTPases through specific phospholipid-dependentsignaling. FEBS Lett. 410, 73–77.

2. Maulik, N., Watanabe, M., Zu, Y-L., Huang, C.-K., Cordis, G. A.,Schley, J. A., and Das, D. K. (1996) Ischemic preconditioningtriggers the activation of MAP kinases and MAPKAP kinase 2 inrat hearts. FEBS Lett. 396, 233–237.

3. Wang, J.-P., Tsao, L.-T., Shen, A.-Y., Raung, S.-L., and Chang,L.-C. (1999) 2-Hydroxymethyl-1-naphthol diacetate (TAC) sup-presses the superoxide anion generation in rat neutrophils. FreeRadic. Biol. Med. 26, 1010–1018.

4. Wilkie, N., Morton, C., Ng, L. L., and Boarder, M. R. (1996)Stimulated mitogen-activated protein kinase is necessary butnot sufficient for the mitogenic response to angiotensin II. A rolefor phospholipase D. J. Biol. Chem. 271, 32447–32453.

5. Atta-ur-Rahman, Harvey, K., and Siddiqui, R. A. (1999) Inter-leukin-8: An autocrine inflammatory mediator. Curr. Pharm.Des. 5, 241–253.

6. Tamura, M., Yoshida, K., and Kataoka, K. (1999) Guanosine59-O-(3-thiotriphosphate)-induced O-2 generation in permeabil-ized neutrophils requires protein kinase C and phospholipase Cbut not tyrosine kinase or phospholipase D. Arch. Biochem.Biophys. 361, 257–263.

7. Kim, J. H., Suh, Y. J., Lee, T. G., Kim, Y., Bae, S. S., Kim, M. J.,Lambeth, J. D., Suh, P.-G., and Ryu, S. H. (1996) Inhibition ofphospholipase D by a protein factor from bovine brain cytosol.Partial purification and characterization of the inhibition mech-anism. J. Biol. Chem. 271, 25213–25219.

310

Neutrophil phospholipase D: inhibition by Rho-GDP dissociationinhibitor and stimulation by small GTPase GDP dissociationstimulator. Methods Enzymol. 256, 246–256.

9. Bonser, R. W., Thompson, N. T., Randall, J. E., Tateson, J. E.,Spacey, G. D., Hodson, H. F., and Garland, L. G. (1991) Deme-thoxyviridin and wortmannin block phospholipase C and D ac-tivation in the human neutrophil. Br. J. Pharmacol. 103, 1237–1241.

0. Naccache, P. H., Caon, A. C., Gilbert, C., Gaudry, M., Roberge,C. J., Poubelle, P. E., and Bourgoin, S. (1993) Inhibition oftyrosine phosphorylation by wortmannin in human neutrophils.Dissociation from its inhibitory effects on phospholipase D. Lab.Inves. 69, 19–23.

1. Kessles, G. C. R., Gervaix, A., Lew, P. D., and Verhoeven, A. J.(1991) The chymotrypsin inhibitor carbobenzyloxy-leucine-tyrosine-chloromethylketone interferes with phospholipase D ac-tivation induced by formyl-methionyl-leucyl-phenylalanine inhuman neutrophils. J. Biol. Chem. 266, 15870–15875.

2. Wang, J.-P., Tsao, L. T., Raung, S. L., Hsu, M. F., and Kuo, S. C.(1997) Inhibition by HAJ11 of respiratory burst in neutrophilsand the involvement of protein tyrosine phosphorylation andphospholipase D activation. Br. J. Pharmacol. 120, 79–87.

3. Takemura, O. S., Banno, Y., and Nozawa, Y. (1997) Inhibition ofN-formylmethionyl-leucyl-phenylalanine-stimulated tyrosinephosphorylation and phospholipase D activation by quercetin inrabbit neutrophils. Biochem. Pharmacol. 53, 1503–1510.

4. Megyeri, P., Pabst, K. M., and Pabst, M. J. (1995) Serine pro-tease inhibitors block priming of monocytes for enhanced releaseof superoxide. Immunology 86, 629–635.

5. Lukowski, S., Lecomte, M.-C., Mira, J.-P., Marin, P., Gautero,H., Russo-Marie, F., and Geny, B. (1996) Inhibition of phospho-lipase D activity by fodrin. An active role for the cytoskeleton.J. Biol. Chem. 271, 24164–24171.

6. Levy, B. D., Fokin, V. V., Clark, J. M., Wakelam, M. J. O.,Petasis, N. A., and Serhan, C. N. (1999) Polyisoprenyl phosphate(PIPP) signaling regulates phospholipase D activity: A ‘stop’signaling switch for aspirin-triggered lipoxin A4. FASEB J. 13,903–911.

7. English, D., and Anderson, B. R. (1974) Granulocytes and mono-nuclear leukocytes on discontinuous density gradients of Ficoll-Hypaque. J. Immunol. Methods 5, 249–252.

8. Bradford, M. M. (1976) A rapid and sensitive method for thequantitation of microgram quantities of proteins utilizing theprinciple of protein-dye binding. Anal. Biochem. 72, 248–254.

9. Davis, L. D., Maglio, J. J., and Horwitz, J. (1998) PhospholipaseD hydrolyzes short-chain analogs of phosphatidylcholine in theabsence of detergent. Lipids 33, 223–227.

0. Van den Bosch, H., De Jong, J. G. N., and Aarsman, A. J. (1991)Phospholipase A2 from rat liver mitochondria. Methods Enzymol.197, 365–373.

1. Smith, D. M., and Waite, M. (1992) Phosphatidylinositol hydro-lysis by phospholipase A2 and C activities in human peripheralblood neutrophils. J. Leukocyte Biol. 52, 670–678.

2. Joseph, D. E., Paul, C. C., Baumann, M. A., and Gomez-Cambronero, J. (1996) S6 kinase p90rsk in granulocyte-macrophage colony-stimulating factor-stimulated proliferativeand mature hematopoietic cells. J. Biol. Chem. 271, 13088.

3. Renold-O’Donnell, E., and Parent, D. (1995) Downregulation ofneutrophil CD34 by opsonized zymosan. Blood 85, 337–342.

4. Megyeri, P., Nemeth, Pabst, K. M., Pabst, M. J., Deli, M. A., andAbraham, C. S. (1999) 4-(2-aminoethyl)benzenesulfonyl fluorideattenuates tumor-necrosis factor-a-induced blood-brain barrieropening. Eur. J. Pharmacol. 374, 207–211.

35. Diatchuk, V., Lotan, O., Koshkin, V., and Wikstroem, P., and

3

phosphoinositide cell content. Biochem. Biophys. Res. Commun.

3


Pick, E. (1997) Inhibition of NADPH oxidase activation by 4-(2-aminoethyl)-benzenesulfonyl fluoride and related compounds.J. Biol. Chem. 272, 13292–13301.

6. Lukowski, S., Mira, J.-P., Zachowski, A., and Geny, B. (1998)Fodrin inhibits phospholipases A2, C, and D by decreasing poly-

311

248, 278–284.7. Nakabo, Y., and Pabst, M. J. (1996) Lysis of leukemic cells by

human macrophages: Inhibition by 4-(2-aminoethyl)-benzene-sulfonyl fluoride (AEBSF), a serine protease inhibitor. J. Leuko-cyte Biol. 60, 328–336.

direct inhibition of in vitro pld activity by 4-(2-aminoethyl)-benzenesulfonyl fluoride

Documents